A watchdog circuit is used to monitor operation of a microprocessor and to reset the microprocessor in the event of a malfunction condition of the microprocessor. Upon detection of the malfunction condition, the watchdog circuit outputs a reset signal to the microprocessor that causes the microprocessor to be reset. The reset operation is intended to return the microprocessor to a known reference state after which the microprocessor resumes operation.
The malfunction condition is typically a result of, though not limited to, any of three causes. The first cause is external electromagnetic interference (EMI), which can cause the power being supplied to the microprocessor to fluctuate, which in turn can cause the microprocessor to operate unpredictably. The second cause, external radiation, such as that found in the atmosphere or outer space, can cause single event effect (SEE) that can also result in erratic operations by the microprocessor. The third cause is unanticipated software errors. Many causes of malfunction conditions can often be cleared by triggering a reset operation of the microprocessor, however, this may not be sufficient to clear certain malfunction conditions.
Single Event Effects (SEEs) are caused by a single, energetic particle. SEEs can be based on soft errors or hard errors. Soft errors can include, for example, a Single Event Upset (SEU), which is usually non-destructive and can be cleared by a reset pulse to the microprocessor. An SEU can appear as s transient pulse in logic or support circuitry, or as a bitflip in a memory cell or register. A hard error can include, for example, a Single Event Latchup (SEL), burnout of power components (e.g., MOSFETS), gate rupture, frozen bits, and noise in CCDs. An SEL that causes a high operating current that exceeds device specifications is potentially destructive. In situations, an SEL can only be cleared by restarting power to the microprocessor, including removing and then restoring power. A reset operation of the microprocessor would not be sufficient.
The watchdog circuit can detect a malfunction condition of the microprocessor, and in response trigger a reset operation of the microprocessor by sending a reset pulse. However, the malfunction condition may still persist, such as when the malfunction condition was caused by an SEL or other hard error. In response to a persistent malfunction condition, the watchdog circuit may hold the reset signal in a persistent actuated state, e.g., by holding the reset signal to a continuous or repeated reset level. The microprocessor is thus maintained in a nonfunctional continuous reset state until maintenance of the hardware is performed.
Such conventional methods and systems have generally been considered satisfactory for their intended purpose. However, there is still a need in the art for responding to malfunction conditions associated with peripheral devices or input/output devices that are coupled to the microprocessor. Additionally, there is still a need in the art for clearing persistent malfunction conditions of the microprocessor or devices coupled to the microprocessor and resuming functional operation. The present disclosure provides a solution for these problems.